Efficacy and safety of colistin

Abstract number: 1733_334

Levcovich A., Lischenski Y., Leon P., Lev S., Hazzan R., Chowers M., Bishara J., Leibovici L., Paul M.

Objectives: The use of colistin decreased substantially in the 1970s following reports of renal failure and neuropathy. The appearance of highly resistant Gram-negative bacteria in hospitals, such as Acinetobacter baumannii, Pseudomonas spp. and Klebsiella spp., rendered colistin a drug of last resort for patients infected with these bacteria. The objectives of this study were to establish the safety and efficacy profile of colistin, in comparison to carbapenems and ampicillin-sulbactam.

Methods: Prospective observational cohort study including all adult patients treated with colistin, imipenem, meropenem or ampicillin/sulbactam with bacteraemia or fulfilling CDC diagnostic criteria for pneumonia, urinary tract infection, meningitis, catheter-related or catheter-associated infections. We compared outcomes for colistin vs. comparator antibiotics. Data were collected at the time of infection presentation with follow-up of 30 days. Mortality was defined as 30-day all-cause mortality. Renal failure was defined as ≥50% increase of creatinine levels from baseline and above 1.4 mg/dL at 2 weeks follow-up.

Baseline patient characteristicsColistinComparatorsP-value
No patients3144 
Female gender14 (45.2%)19 (43.2%)0.86
Age (mean, SD)68 (±16)57 (±20)0.029
  Internal medicine12 (38.7%)14 (31.8%)0.29a
  Surgery4 (12.9%)14 (31.8%) 
  ICU15 (48.4%)16 (36.4%) 
Baseline functional status (full activity)17 (54.8%)31 (70.5%)0.17
Central line10 (32.3%)19 (43.2%)0.34
Mechanical ventilation18 (58.1%)25 (56.8%)0.91
Recent invasive procedure18 (58.1%)29 (65.9%)0.49
Charison score (median, range)2 (1–6)1 (1–7)0.31
McCabe score
  117 (54.8%)32 (72.7%)0.11b
SOFA score (median, range)4 (1–14)4 (1–14)0.68
Chronic renal failure10 (32.3%)11 (25%)0.49
Albumin g/dL (mean, SD)2.2 (±0.46)2.4 (±0.66)0.33
WBC K/micl (mean, SD)16.587 (±9.6)14.294 (±8.4)0.28
Bacteraemia12 (38.7%)13 (29.5%)0.41
aFor ICU vs. others; bfor 1 vs. others.

Results: Data were collected between March to October 2006 in 2 medical centres in Israel; 31 patients were treated with colistin and 44 with comparator antibiotics (meropenem 12, imipenem 15, ampicillin-sulbactam 17). All infections were healthcare associated (median time in hospital before infection 17 days, 0–34). Prior antibiotic treatment was administered to 27 patients with colistin vs. 38 patients with comparator antibiotics. Treatment was administered for microbiologically documented infections in all but 2 patients. Acinetobacter, Klebsiella and Pseudomonas spp. were implicated in 56%, 56% and 31% of infections, respectively. Colistin was administered as single drug to 27/31 colistin-treated patients; bacteria were susceptible only to colistin in 23/3. Baseline patient characteristics and outcomes are shown in the table. Mortality was significantly higher with colistin vs. comparators, 48% vs. 20%, p = 0.01. A multivariate analysis including 75 patients and the variables age, McCabe score, independent functional status and treatment arm as covariates, revealed that colistin remained significantly associated with 30-day mortality (p = 0.03).

Conclusions: Treatment with colistin was associated with a significantly higher mortality rate at 30 days and a non-significantly higher rate of renal failure development. Higher-risk patients were treated with colistin.

Session Details

Date: 31/03/2007
Time: 00:00-00:00
Session name: European Society of Clinical Microbiology and Infectious Diseases
Location: ICC, Munich, Germany
Presentation type:
Back to top